<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv=Content-Type content="text/html; charset=us-ascii">
<meta name=Generator content="Microsoft Word 12 (filtered medium)">
<!--[if !mso]>
<style>
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style>
<![endif]-->
<style>
<!--
/* Font Definitions */
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman","serif";}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
p
{mso-style-priority:99;
mso-margin-top-alt:auto;
margin-right:0in;
mso-margin-bottom-alt:auto;
margin-left:0in;
font-size:12.0pt;
font-family:"Times New Roman","serif";}
span.EmailStyle18
{mso-style-type:personal;
font-family:"Calibri","sans-serif";
color:#1F497D;}
span.EmailStyle19
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page Section1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.Section1
{page:Section1;}
-->
</style>
<!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=EN-US link=blue vlink=purple>
<div class=Section1>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>For list archival purposes, copy of an e-mail that I just sent
John F. w/o the attachment.<o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>Jean-Paul<o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<div>
<div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in'>
<p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b><span
style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'> Watson, Jean-paul <br>
<b>Sent:</b> Friday, February 27, 2009 8:20 PM<br>
<b>To:</b> 'John J Forrest'<br>
<b>Cc:</b> Jonathan Eckstein; Phillips, Cynthia A<br>
<b>Subject:</b> RE: [Cgl] Issue/potential bug with CglGomory cuts<o:p></o:p></span></p>
</div>
</div>
<p class=MsoNormal><o:p> </o:p></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>Hi John,<o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>Thanks for the suggestion – I had thought we were
up-to-date, and spent time this PM making sure we were (after some code fixes
due to deprecated COIN-related APIs). Specifically, I brought in the latest
coin-cbc stable branch (2.2). <o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>The fix did correct the PP08 problem, but there remain other
instances where cuts are not generated. I’m attaching the smallest remaining
one (out of about 15), which is derived from set1ch. <o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>Given that you aren’t filtering in the old style, I have
even less of an idea why cuts would not be generated in this case. <o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>I can send additional “break” instances if you want,
but figured the current one-at-a-time paradigm is making for good progress thus
far.<o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'>Jean-Paul<o:p></o:p></span></p>
<p class=MsoNormal><span style='font-size:11.0pt;font-family:"Calibri","sans-serif";
color:#1F497D'><o:p> </o:p></span></p>
<div style='border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in'>
<p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b><span
style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>
cgl-bounces@list.coin-or.org [mailto:cgl-bounces@list.coin-or.org] <b>On Behalf
Of </b>John J Forrest<br>
<b>Sent:</b> Friday, February 27, 2009 1:39 PM<br>
<b>Cc:</b> Cgl@list.coin-or.org<br>
<b>Subject:</b> Re: [Cgl] Issue/potential bug with CglGomory cuts<o:p></o:p></span></p>
</div>
<p class=MsoNormal><o:p> </o:p></p>
<p class=MsoNormal style='margin-bottom:12.0pt'><br>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>Jean-Paul,</span>
<br>
<br>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>Firstly, I
would suggest you upgrade your version of Cgl from 0.52 to 0.53. The
goodCuts code you complained about was removed from 0.53 and the relaxation
tightened a bit. So with 0.53 I get changes with both matrices.</span> <br>
<br>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>I am looking at
relaxation in trunk - at present the code runs slower if I tighten too much.</span>
<br>
<br>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>John</span> <o:p></o:p></p>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><img width=100 height=1 id="_x0000_i1025"
src="cid:image001.gif@01C99918.CA879DC0"><o:p></o:p></p>
</td>
<td width="100%" style='width:100.0%;padding:.75pt .75pt .75pt .75pt'>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td width="100%" valign=top style='width:100.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>RE:
[Cgl] Issue/potential bug with CglGomory cuts</span></b><o:p></o:p></p>
</td>
</tr>
</table>
<p class=MsoNormal><o:p> </o:p></p>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Arial","sans-serif";
color:#E26200'>Watson, Jean-paul </span></b><o:p></o:p></p>
</td>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:10.0pt;font-family:"Arial","sans-serif";
color:#8F8F8F'>to:</span> <o:p></o:p></p>
</td>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>John
J Forrest</span> <o:p></o:p></p>
</td>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal align=right style='text-align:right'><span
style='font-size:7.5pt;font-family:"Arial","sans-serif"'>02/27/2009 11:37
AM</span><o:p></o:p></p>
</td>
</tr>
</table>
<p class=MsoNormal><o:p> </o:p></p>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td style='padding:.75pt .75pt .75pt .75pt'>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:7.5pt;font-family:"Arial","sans-serif";
color:#8F8F8F'>Cc:</span> <o:p></o:p></p>
</td>
<td width="100%" style='width:100.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:7.5pt;font-family:"Arial","sans-serif"'>"Cgl@list.coin-or.org"</span><o:p></o:p></p>
</td>
</tr>
</table>
</td>
<td style='padding:.75pt .75pt .75pt .75pt'></td>
</tr>
</table>
<p class=MsoNormal><o:p></o:p></p>
</td>
</tr>
</table>
<p class=MsoNormal style='margin-bottom:12.0pt'><o:p> </o:p></p>
<div class=MsoNormal align=center style='text-align:center'>
<hr size=2 width="100%" align=center>
</div>
<p class=MsoNormal><br>
<br>
<br>
<span style='font-size:10.0pt;font-family:"Calibri","sans-serif";color:#004080'>Sure
– I’ve attached two mps files (one for which cuts will be
generated, one for which they won’t) and a simple test driver based on
some Cgl example. </span><br>
<span style='font-size:10.0pt;font-family:"Calibri","sans-serif";color:#004080'> </span>
<br>
<span style='font-size:10.0pt;font-family:"Calibri","sans-serif";color:#004080'>Jean-Paul</span>
<br>
<span style='font-size:10.0pt;font-family:"Calibri","sans-serif";color:#004080'> </span>
<br>
<b><span style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>From:</span></b><span
style='font-size:10.0pt;font-family:"Tahoma","sans-serif"'>
cgl-bounces@list.coin-or.org [mailto:cgl-bounces@list.coin-or.org] <b>On Behalf
Of </b>John J Forrest<b><br>
Sent:</b> Friday, February 27, 2009 7:40 AM<b><br>
Cc:</b> Cgl@list.coin-or.org<b><br>
Subject:</b> Re: [Cgl] Issue/potential bug with CglGomory cuts</span> <br>
<br>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
There was an accuracy issue with Gomory cuts. It is very likely that the
correction was over drastic.</span> <br>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
If you give me some code, I can try and correct the correction.</span> <br>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
John Forrest</span> <o:p></o:p></p>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td width="8%" style='width:8.0%;padding:.75pt .75pt .75pt .75pt'></td>
<td width="91%" style='width:91.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><o:p> </o:p></p>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td width="100%" valign=top style='width:100.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>[Cgl]
Issue/potential bug with CglGomory cuts</span></b><o:p></o:p></p>
</td>
</tr>
</table>
<p class=MsoNormal><br>
<o:p></o:p></p>
<p><o:p> </o:p></p>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td width="33%" style='width:33.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Arial","sans-serif";
color:#E26200'>Watson, Jean-paul </span></b><o:p></o:p></p>
</td>
<td width="5%" style='width:5.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:10.0pt;font-family:"Arial","sans-serif";
color:#8F8F8F'>to:</span> <o:p></o:p></p>
</td>
<td width="32%" style='width:32.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:10.0pt;font-family:"Arial","sans-serif"'>Cgl@list.coin-or.org</span>
<o:p></o:p></p>
</td>
<td width="28%" style='width:28.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal align=right style='text-align:right'><span
style='font-size:7.5pt;font-family:"Arial","sans-serif"'>02/26/2009 10:37
PM</span><o:p></o:p></p>
</td>
</tr>
</table>
<p><br>
<o:p></o:p></p>
<p><o:p> </o:p></p>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td width="98%" style='width:98.0%;padding:.75pt .75pt .75pt .75pt'>
<table class=MsoNormalTable border=0 cellpadding=0 width="100%"
style='width:100.0%'>
<tr>
<td width="22%" style='width:22.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:10.0pt;font-family:"Arial","sans-serif";
color:#8F8F8F'>Sent by:</span> <o:p></o:p></p>
</td>
<td width="77%" style='width:77.0%;padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><b><span style='font-size:10.0pt;font-family:"Arial","sans-serif";
color:#E26200'>cgl-bounces@list.coin-or.org</span></b> <o:p></o:p></p>
</td>
</tr>
<tr>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:7.5pt;font-family:"Arial","sans-serif";
color:#8F8F8F'>Cc:</span> <o:p></o:p></p>
</td>
<td style='padding:.75pt .75pt .75pt .75pt'>
<p class=MsoNormal><span style='font-size:7.5pt;font-family:"Arial","sans-serif"'>"Carr,
Robert D", "Hart, William E"</span><o:p></o:p></p>
</td>
</tr>
</table>
</td>
<td width="1%" style='width:1.0%;padding:.75pt .75pt .75pt .75pt'></td>
</tr>
</table>
<p class=MsoNormal><o:p></o:p></p>
</td>
</tr>
</table>
<p><br>
<o:p></o:p></p>
<p class=MsoNormal align=center style='text-align:center'><o:p> </o:p></p>
<div class=MsoNormal align=center style='text-align:center'>
<hr size=2 width="100%" align=center>
</div>
<p class=MsoNormal><br>
<br>
<br>
<span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
Dear CGL’ers:</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
I’ve been working with Cgl in the context of our PICO parallel MIP
solver, and ran into an issue and/or bug that I want to raise with the Cgl
development team.</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
The problem context is “pp08” from either miplib3 or miplib2003; I
forget exactly which at the moment. If I solve the root LP relaxation and then
attempt to find CglGomory cuts, several are found and correctly applied. However,
if I modify the objective coefficients only – we’re working with
feasibility pump-like heuristics – and solve pp08, CglGomory isn’t
able to identify any cuts. For us, this is a major issue because we are using
Gomory cuts to escape local minima in our heuristic, and they should (to a
first order, ignoring numerical issues) always exist. It also happens on
roughly 1/3 of the miplib2003 instances, which is preventing us from completing
a reasonable set of experimental tests.</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
I have replicated this behavior with a pure Clp/Cgl test driver, so I’m
certain it isn’t related to the PICO use of Cgl/Clp. I can provide the
Clp-based test driver and problem instances if anyone is interested.</span> <span
style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
Digging into the CglGomory.cpp cut generation code, it appears that cuts are
being found, but then deemed “unworthy” and then thrown away. The
relevant piece of logic is:</span> <span style='font-size:10.0pt;font-family:
"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
if
(number<limit||!numberNonInteger) <br>
{</span> <span
style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
bool goodCut=true;</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
bounds[1]=rhs;</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
if
(number>50&&numberNonInteger)</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
{</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
bounds[1] = bounds[1]+1.0e-6+1.0e-8*fabs(rhs); // weaken</span> <span
style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
}</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
if
(number>5&&numberNonInteger&&relaxation>1.0e-20) <br>
{</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
relaxation *= fabs(rhs);</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
//printf("relaxing rhs by
%g\n",CoinMin(relaxation,1.0e-3));</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
bounds[1] = bounds[1]+CoinMin(relaxation,1.0e-3); // weaken</span>
<span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
#if 1</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
//bounds[1] = bounds[1]+relaxation*fabs(rhs); // weaken</span> <span
style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
// may be better to thow away</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
if (relaxation>1.0e6)</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
{</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
goodCut=false;</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
}</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
#endif</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
}</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
The generated cuts fall prey to the “goodCut=false” assignment,
which is due to the condition number (the variable “relaxation”)
exceeding the indicated threshold. Typically, I am seeing cuts with number=6 or
number=7, with numberNonInteger>=1, and a large relaxation value. <br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
After some internal discussion, we realize we don’t understand the logic
behind this culling procedure. For example, we’d like some insight into:</span>
<span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
1. </span><span style='font-size:10.0pt;font-family:
"Calibri","sans-serif"'>The intuition behind scaling the relaxation by the
fabs(rhs) factor</span> <span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
2. </span><span style='font-size:10.0pt;font-family:
"Calibri","sans-serif"'>Why the relaxation values coming into this fragment are
so large (e.g., on the order of 10^50 for the modified problem), when in CPLEX they
are quite small (on the order of 100-10000 for the same problem).</span> <span
style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
3. </span><span style='font-size:10.0pt;font-family:
"Calibri","sans-serif"'>The intuition behind triggering of culling logic only
when “numberNonInteger” is > 0.</span> <span style='font-size:
10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
The good news is that by commenting out the “goodCut=false” line,
we run fine. But that isn’t the right answer either. Perhaps a
user-specified option might be, e.g.,
“generate-without-culling=true”. <br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
I am also wondering if updating this logic might improve the proving-optimality
phase of many Cgl-based MIP solvers, including PICO. We have observed that it
seems to under-cut relative to CPLEX.</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
There are many more details, but the intent is to start a discussion. Any help
is greatly appreciated on our end!</span> <span style='font-size:10.0pt;
font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
Jean-Paul</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
--</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
Dr. Jean-Paul Watson</span> <span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
Discrete Math and Complex Systems Department</span> <span style='font-size:
10.0pt;font-family:"Arial","sans-serif"'><br>
Sandia National Laboratories</span> <span style='font-size:10.0pt;font-family:
"Arial","sans-serif"'><br>
P.O. Box 5800, MS 1318</span> <span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
Albuquerque, NM 87185-1318 USA</span> <span style='font-size:10.0pt;font-family:
"Arial","sans-serif"'><br>
(505) 845-8887</span> <span style='font-size:10.0pt;font-family:"Arial","sans-serif"'><br>
jwatson@sandia.gov</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span> <span style='font-size:10.0pt;font-family:"Calibri","sans-serif"'><br>
</span><span style='font-size:10.0pt;font-family:"Courier New"'>_______________________________________________<br>
Cgl mailing list<br>
Cgl@list.coin-or.org<br>
http://list.coin-or.org/mailman/listinfo/cgl[attachment
"somecuts.mps" deleted by John J Forrest/Watson/IBM] [attachment
"cuttest.cpp" deleted by John J Forrest/Watson/IBM] [attachment
"nocuts.mps" deleted by John J Forrest/Watson/IBM] </span><o:p></o:p></p>
</div>
</body>
</html>